In-Vehicle Radar Calibration Using Odometry Error Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Radar devices in vehicles suffer from inaccuracies due to optical axis deviations and errors in vehicle speed detection, leading to incorrect object detection and vehicle positioning.
Innovation Solution
An in-vehicle sensor device comprising a radar, odometry sensor, and processing unit that estimates and updates position information parameters, mounting angles, and detection errors using an extended Kalman filter to improve object recognition accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If the radar device uses odometry sensor data (vehicle speed and angular velocity) to detect straight motion and calculate optical axis deviation, then the device can perform automated correction, but detection accuracy decreases due to errors in the odometry sensor measurements
Solution Approach 1:
The system uses feedback from multiple sensors (radar, gyro, odometry) to continuously monitor and correct optical axis deviation. The radar detection results are fed back to verify whether the vehicle is actually moving straight, and this feedback is used to correct the optical axis deviation amount calculated from odometry data, creating a closed-loop correction system that improves accuracy while maintaining automation.
Solution Approach 2:
The patent combines multiple detection methods into a unified system: radar-based object detection, gyro-based straight motion detection, and odometry-based optical axis deviation calculation are merged. By integrating these different sensing approaches, the system leverages the strengths of each sensor while compensating for their individual weaknesses, achieving both automation and accuracy.
2Device complexity
If the radar device relies solely on odometry sensor data for correction, then the system remains simple, but incorrect detection occurs when the vehicle does not move straight despite error-containing detection information
Solution Approach 1:
The radar detection function serves as an intermediary verification mechanism. Instead of relying directly and solely on odometry sensor data, the system uses radar-based object detection as an intermediate step to verify whether the vehicle is actually moving straight. This intermediary check prevents incorrect corrections by filtering out cases where odometry data may be erroneous.
Solution Approach 2:
The correction amount is made dynamic rather than static. The system adjusts the optical axis deviation correction based on real-time radar detection results, making the correction adaptive to actual driving conditions. This dynamic approach allows the system to maintain reliability by adjusting corrections based on actual detected straight motion rather than relying on potentially erroneous fixed odometry data.
Data Source
AI summary
An in-vehicle sensor device has an active sensor, an odometry sensor and a processing part. The processing part calculates estimated detection values of a stationary object by using position information parameters, a mounting angle of the active sensor, and a detection error of the odometry sensor. The position information parameters specify a relative position relationship between the stationary object and the active sensor. The processing part updates the position information parameters, the mounting angle, and the detection error simultaneously on the basis of a difference between the estimated detection value which has been calculated, and the position values of the stationary object detected by the active sensor.